Airline passenger satisfaction analysis: Customer analysis

● To provide marketing insights for the client to boost customer satisfaction, particularly focus on disloyal customers.

● What are the key pain points of disloyal customers? How to satisfy them?

● Kaggle

Analysis of Approach the customer base

Customer Segmentation

Analysis of the key pain points

Prediction of customer overall satisfaction

As expected, disloyal customers show high dissatisfaction level.

Majority of them are female in their 20s who often go on business trips.

Dissatisfied customers tend to fly short distances.

Seemingly, online boarding and inflight entertainment affect their satisfaction level.

So, the company needs to boost disloyal customers’ satisfaction level first… However,

given their heterogeneity, we figured that standardized approach will not be effective.

We first segmented disloyal customers into four groups. Each group must be suffering from different pain points.

We ran logistic regression model on each segment to identify their key pain points.

Each segments' features are shown below:

After we ran the logistic regression, we now understand the different persuasion. We suggest the differentiated marketing solutions for each segment. Suggestion are written with following graph:

In this case we apply two machine learning methodologies: Logistic regression and K-means clustering.

Logistic regression measures service quality characteristics that contribute to customer satisfaction thus helping the airline company to make informed decisions on different segmentations.

Clustering helped use to segment customers in to different groups according to their response of service. It increase the efficiency by focus on specific solutions to increasing the satisfaction on each cluster customers.